All-in-one ready to use package (maybe with the use of Arduino) (Total evaluation: 65)
5
3
5
Surface heating of the flasks with the use of resistances. (e.g. kitchen stove)(Total evaluation: 44)
3
2.6
3.2
Usage of heating tape (Total evaluation: 51)
4
2.8
3.4
Chamber
Transparent benchtop door, controls down
4.6
3.2
4.8
Sheet metal bending
5
4.4
4.8
Plexiglass, Aluminium profile columns, door, silicone for isolation)
5
4.4
4.8
Lifting cover
5
3.6
3
Box chamber with an air gap for maximum isolation
3
3
3.4
Welded sheets of metal or plastic, with a transparent cover/door in front
4
2.8
4
Conceptual Design Εvaluation
Motion System:
Magnetic Stirrer (Total score 24/22/24) for the two Erlenmayer flasks, Linear Shaker (Total Score 20/14/21) for the ten test tubes.
The magnetic stirring system poses no difficulty in construction, due to the simplicity of the design, the availability and the low cost of the fans and the magnetic parts. Complementarily to this system, the linear shaker system will be used for the shaking of the test tubes, as the low remaining load of the test tubes can be easily manipulated by a stepper motor and a simple linear reciprocating mechanism. A draft 3D sketch of the shaking/stirring system is presented below:
Temperature Control System:
A heating element (resistor) and a fan are combined to provide enough hot air flow to keep the system's temperature steady (32 or 37 Celcius, depending on the substance).
Temperature control could be achieved either by controlling the current through the resistor or by adjusting the rotating speed of the fan and therefore the air flow. The real challenge and main issue is to keep the resistor's temperature at safe levels to avoid inflammation (by providing enough air flow to cool the resistor through convection).
This system is relatively cheap, with most elements being easily accessible (e.g. inside a hairdryer).
Chamber choice: Plexiglass or bent sheets of metal, Transparent door, Silicone for isolation.
Most of the required materials are already in our disposal.
Sheets of Plexiglass can be processed with laser cut.
Aluminium profile columns can be purchased from local stores. In case of difficulties in finding the desired products, delayed shipments or other problems, we could simply use bolts or welding methods for assembling.
Silicone can be used for thermal insulation.
Special support parts for the Temperature Control and Motion systems can be 3D printed with ABS material using our laboratory’s 3D printers.
Requirements:
- 50 mm stroke
- 150 strokes/min shaking frequency
- 1 kg maximum shaking table load
For the definition of the crank, rod lengths (l1, l2 respectively):
- l1=25 mm, i.e. half the stroke length
- For the length of the rod, various values where tested, subjected to the kinematics restriction (l2>=l1). The values of the velocity in each case are shown below. For (l1=l2) the velocity [m/s] - crank angle [rad] function plot (figure 3.1) shows that the test tube table moves with high velocities only between {-pi/2+2·k·pi, pi/2+2·k·pi}, where k=0,1,2,... . That creates high inertial acceleration resulting in impulsive loads exerted at the joints and the motor.
Figure 2.1: velocity=function(cam_angle) for l1=l2.
With a little research [3.1] in combustion engine crank-rod mechanisms, we deduce that some standard values of the rod length resulting in reduction of the above described phenomenon, are 3-5 times the length of the crank. Chosing a rod length three times the crank length (l2=75 mm), results in a much more gentle velocity curve (Figure 3.2) and finite values for the acceleration curve (Figure 3.3).
Figure 2.2: velocity=function(cam_angle) for l1=3·l2. The max velocity value was calculated as 0.207057 [m/s].
Figure 2.3: acceleration=function(cam_angle) for l1=3·l2. The max acceleration value was calculated as 1.11604 [m/s^2].
From the maximum value of the accelerarion, we can calculate the maximum inertial force exerted at the table of total weight of 1 kg, being Fmax=1.11604 [N]. Multiplying the maximum force with the maximum velocity,results in the ideal maximum power consumption being Pi,max=0.2311 [W]. To be able to sustain a reciprocating frequency of 150 strokes/min in this mechanism, a steady angular crank velocity of n=75 [rpm] is needed. We chose a fitting dc motor for this purpose {n_max=72 rpm, P_max=7 W, V=12V}. After some early mechanical strength calculations the following crank-rod mechanism was designed:
Figure 2.4: Crank-rod linear shaking mechanism.
List of Components:
DC Motor 75 rpm
Crank-Rod
2 bent corners
2 flanged ball bearings MF85ZZ
2 axel
2 linear ball bearings
2 custom made threaded studs
1 adjustable linear bearing nest
Air Heating System
Most of the commercially available operational shaker-incubators do not clarify how the temperature control system works. However, there are basically two options:
1) Use of a PID (or similar) controller. One could also try to model the thermal system (resistor, capacities, sources, etc) but it is doubtful that this estimation would be anywhere near reality due to the complexity of such systems.
2) Activation of the heating system (in our case, the fan and the resistor) when the temperature falls below minimum (in our case 32 or 37 Celcius). The system operates until temperature reaches maximum (defined by the user) and then it is deactivated. When the temperature reaches its minimum allowed value again (because of thermal losses), the system is activated again, and so on.
The main difference between the two methods is, as implied above, that the first is constantly adjusting the temperature so that is does not diverge significantly from the set value. However, it is more complex than the second and it requires the definition of gains. In the second case we accept that there is going to be an important fluctuation of the temperature's value, but we end up with a simple operating system in return. We decided to use the second method of control, regarding the reasons mentioned above. There will be one thermocouples,
The selection of both the fan and the resistor is of great importance in this application. The reason for this is that the air flow that the fan produces needs to absorb the heating power of the resistor, otherwise the resistor will overheat and eventually melt. This, of course, will lead the whole system to failure.
The resistor that will be used in this application can produce 1500 W and it is made out of nichrome whose melting temperature is 1400 oC. The fan that was chosen for this project can provide 145 m3/h of air flow. The following calculations will show that system above can be balanced so that the resistor's temperature does not exceed nichrome's melting point.
In a first attempt to get a good estimation of how the heat transfer phenomena will affect the temperature of the resistance, the following assumptions were made. The first assumption was that the main phenomenon of heat transfer, in this application, is convection. And the second one was that the shape of the resistance can be modeled by a cylinder of the same diameter and length as the actual resistance. The second assumption, although not true, was made in order to approximate the heat convection factor. The heat convection factor is affected by many variables such as the relative speed between the liquid that moves around the resistance (in this case air) and the resistance itself, the kinematic viscocity of the liquid, the diameter of the resistance and the type of the flow (direction in which the flow approaches the resistance. So, it is easy to understand that the heat convection factor is very hard to calculate. However, the heat transfer for a flow around a cylinder is a well studied phenomenon and thus the equations that describe it were available to us.
Having all the above in mind , the following equations can be written:
P_resistor=Q ̇_convection⟹ 1500W=m ̇_air∙C_p∙ΔT⟹
So, it is possible to calculate the air flow needed to achieve the desired air temperature int the incubator which is approximately 40oC:
m ̇_air=1500W/(1005 J/(kg∙°C)∙(40-15)℃)=0.059 kg/s⟹
Therefore, the velocity of the incoming air can be found from the following equation:
v_air=m ̇_air/(ρ∙Α_fan )=(0.059 kg/s)/(1.2 kg/m^3 ∙π∙((0.115 m)/2)^2 )=14.87 m/s
For this velocity, it is possible to estimate the convective heat transfer factor:
h=100 W/(m^2∙K)
Since all the above is found, the temperature of the resistor T_s is calculated:
Q ̇_convection=A_resistor∙h∙(T_s-15°C)⟹ T_s=1210℃
So, it is shown that the resistor will operate at a temperature which is well underneath its melting point and thus the selection of the fan is suitable. However, the model of the cylinder that we used in this approach cannot describe the reality in a very satisfactory way, since the temperature found was well above the expected values.
For this reason, a second approach was attempted. In this approach, the program Ansys Transient Thermal was used to get a more accurate value of the temperature of the resistance as well as the time in which this temperature is achieved. In order to get a better estimation, a better model of the resistance was needed. In that spirit, the shape of a cone with a hole around its axis was adopted. The assumption that convection is the major phenomenon taking place, though, was not altered. After the program was used the following results were introduced:
2.5 : Resistanse steady state temperature
2.6: Temperature to time
So, in these diagrams we can see that the temperature of the resistances is stabilized at 469.85 oC after 50 seconds. These results are much more appropriate to describe the result we expect and they show that the resistance is not in danger of overheating. However, it is important to notice that the inputs of the program were the power of the resistance and the heat convection factor. For the second input we used the value of the convection factor that we calculated in the previous approach. That means that this is only a good approximation and the necessity of an experiment that verifies our results cannot be overwritten.
Because the fan was larger than the resistor (120 mm diameter vs 40 mm), and because the computer fan used is not designed to provide axial flow, it is necessary to use a "tube" in order to focus the air flow on the resistor. Running a simulation on ANSYS CFX we found that:
2.7: Pressure contour after fan
2.8: Overwiew
Pressure right after the fan measures about to 650 Pascal which is equal to 66.281 mm H2O. However, our fan, according to the pressure-air flow curve, can provide up to 8 mm of pressure.
2.9: Pressure - air flow curve (50 Hz)
We could therefore conclude that this fan is not appropriate for this kind of application. However an experiment was conducted. We constructed a frame from a carton box, like the one we simulated, only square, and we activated the fan. There was not as much air flow as without the frame, but when the resistor was fed with current, the temperature quickly rose to 50 degrees, more than enough. The only problem was that temperature decreased afterwards as quickly as it rose, requiring activation of the resistor through the relay every 5-10 seconds to keep it about 35 degrees. As a final conclusion, we should keep in mind that this system works well for small amounts of time, without any danger of inflammation, with the drawback that it has to be activated about every 20 seconds (if we consider that the actual construction will be better insulated than our carton frame).
Electric subsystem
The main challenge concerning the design of the electrical subsystem is the proper isolation between the AC and DC components, as well as their protection from excessive currents. Special care should be taken to protect the Arduino board, which is the most fragile and expensive component.
Considering our system should have a single plug to power the motors, the Arduino board, the resistor, the fan and every other electronic or electrical component, it is clear that the use of a transformer is imperative. Input current (230V/50Hz) supplies the resistor an its fan before transformation to +12VDC, which is used to supply the motion system's motors and the Arduino.
Other components include a temperature sensor (DS18B20, http://grobotronics.com/temperature-sensor-ds18b20-el.html ) requiring +5V supply and a relay (https://www.parallax.com/product/27115) to control the resistor. A relay could also be used to switch the fan on and off, but we chose to keep the fan always functioning to provide better air circulation, which is another system requirement.
Safety precautions are taken to protect the system's integrity. There is a fuse in the main power adaptor as well as a master switch which can be used to shut down the entire system in case of malfunction. Measures should also be taken to ensure that the resistor can not be supplied unless the fan is functioning, to avoid inflammation. In the schematics that follow, switch A ensures that this does not happen. Switch B is used to deactivate temperature control in case it is required to stir something in room temperature.
Last, concerning current draw, it is calculated to about 0.5-0.6 A for the DC subsystem and about 7 A for the AC subsystem, plus the transformer AC current draw. The fuses should be chosen accordingly, also taking into account higher currents that occur when motors start rotating.
The circuit is pictured below:
2.10: Complete electric circuit
Casing:
The main goals of the chamber's design are the following:
Steady construction
Minimum amount of machining processes
Easy to assembly and disassembly in case of emergency or failure
Thermal isolation of the chamber
Liquid isolation of the electric parts
We came up with two main designs for the chamber:
1st design
In this solution, the only material and machining process necessary is laser cut of PMMA/Plexiglass material.
All the construction is assembled with screws, brackets and geometric special formations.
All the Motion System parts are placed on two slabs that are assembled together with four threaded rods.
After that, we can assemble the rest of the chamber with simple M5 screws and nuts.
In the back side, all the Thermal System Components are placed. Then, the cover is assembled using two simple hinges.
The last piece of the chamber is placed using the geometry formation showed in the next picture and can be screwed with 2 threaded rods as seen below:
This construction has the benefit that it is extremely easy to assemble, using just a screwdriver and a wrench but has the downside that it's heavy, around 10kilos.
2nd design
In the second solution, we replaced the most PMMA parts with two steel sheets. Each sheet's thickness is 0,5mm. This reduced the total weight of the chamber by 40% and made the assembly procedure even easier and quicker.
The Motion and Thermal system controls are placed on the front face of the chamber.
The Thermal System components are placed inside the back of the chamber.
Under the chamber's bottom there are 4 rubber pads that can absorb the vibrations caused by the Motion and Thermal System.
The construction includes a microswitch connected with our arduino and is placed on the left side of the chamber.
The switch's purpose is to stop the stirring function when the cover is open.
Chamber's Total Cost:
The total cost of the brackets/pads/screws/nuts/hinges and handle is ~15euros
The plexiglass slabs are already in our disposal.
The price of the bent steel sheet is still unknown. It depends on the machine shop's offer.
Experimental setup:
1 acrylic base
4 cylindrical magnets
1 screw
Rotational motion of a drill
1 Erlenmeyer flask with 100 ml water
1 magnetic capsule
Experimental conclusions:
A magnetic stirrer with the existing magnetic setup will function well in 300 rpm, given that the rotational speed will be increasing gradually. The magnetic capsule rotates due to the magnetic field created by the four magnets, creating the needed vortex for the air to mingle with the water.
Project Info:
Objective:
Design and fabricate a shaking incubator for culture of bacteria and yeast.Team Members:
1.Γαρυφαλλίδης Σπυρίδων2.Παππάς Ιωάννης3.Δαλλάς Σπυρίδων4.Μαργέλης Νικόλαος5.Παναγιώτου ΝίκοςTeam Contact:
Δαλλάς Σπυρίδων (spyro.d.mechs@gmail.com)Mentor:
Tzeranis Dimitrios (tzeranis@gmail.com)Consultants:
Kanakaris Georgios (gkanak84@gmail.com)Polesiouk Alexandros
Instructions for members:
Milestone 1: Preliminary design
Design Specifications
Flask Capacity: 2 Erlenmeyer 500mL flasks, 10 test tubesShaking Range: 250-300 rpm
Orbit diameter: 30 mm
Temperature Range: 32-37 Celcius
Safety Factor: >2
Cost: Minimum
Key Functions
Mechanical Motion Transition system for shaking purposes.Temperature Control system.
Chamber construction.
Conceptual Design Brainstorming
To acquire the desired reciprocal movement 150 strokes/min
a mean linear velocity of 75 mm/sec is required.
-Spectral lab instruments: Linear Motion Shaker
-Straight axis solution
-Bent axis solution
τ < = r^2*ω^2*m
where r is orbital radius, ω is angular speed in [rad/sec] and m is the total mass of the platform and the flasks.
-Stuart shaking incubator
-Patent: Incubating orbital shaker US8393781B2
-Drive solution (gears, belt) cost
-Cam misalignment
- Transparent benchtop door ( free angle: yaw)
-Non transparent box
-F & K Scientific: KS 4000 Control Shaking Incubator
-Acrylic sides for transparency
-Silicone fastening
-Acrylic cover for transparency
Professional-looking design
Lightweight
Easy construction
2: http://www.makeuseof.com/tag/make-your-own-temperature-controller-with-an-arduino/
3: http://www.directindustry.com/prod/ika/product-28268-1152623.html
4: http://www.labtech.in/main.php?cPath=21_71&crdID=xvfodupt
5: http://www.instructables.com/id/How-to-Make-a-Cheap-Portable-Magnetic-Stirrer/
6: https://www.youtube.com/watch?v=XylXujEpuEw
Figure 1.1: Orbital-Rotational Mechanisms
Figure 1.2: Orbital mechanism with three cams
Figure 1.3: Cheap magnetic stirrer
Solution Evaluation Table
(1-5)
(1-5)
(1-5)
Conceptual Design Εvaluation
Motion System:
Magnetic Stirrer (Total score 24/22/24) for the two Erlenmayer flasks, Linear Shaker (Total Score 20/14/21) for the ten test tubes.
The magnetic stirring system poses no difficulty in construction, due to the simplicity of the design, the availability and the low cost of the fans and the magnetic parts. Complementarily to this system, the linear shaker system will be used for the shaking of the test tubes, as the low remaining load of the test tubes can be easily manipulated by a stepper motor and a simple linear reciprocating mechanism. A draft 3D sketch of the shaking/stirring system is presented below:Temperature Control System:
A heating element (resistor) and a fan are combined to provide enough hot air flow to keep the system's temperature steady (32 or 37 Celcius, depending on the substance).Temperature control could be achieved either by controlling the current through the resistor or by adjusting the rotating speed of the fan and therefore the air flow. The real challenge and main issue is to keep the resistor's temperature at safe levels to avoid inflammation (by providing enough air flow to cool the resistor through convection).
This system is relatively cheap, with most elements being easily accessible (e.g. inside a hairdryer).
Chamber choice: Plexiglass or bent sheets of metal, Transparent door, Silicone for isolation.
Most of the required materials are already in our disposal.
Sheets of Plexiglass can be processed with laser cut.
Aluminium profile columns can be purchased from local stores. In case of difficulties in finding the desired products, delayed shipments or other problems, we could simply use bolts or welding methods for assembling.
Silicone can be used for thermal insulation.
Special support parts for the Temperature Control and Motion systems can be 3D printed with ABS material using our laboratory’s 3D printers.
Milestone 2: Detailed design
Erlenmeyer Flask Stirring System
1) Use of 2x [ 300RPM, 12V DC-Motors ] for Magnetic Stirring
http://grobotronics.com/standard-gearmotor-303-rpm-3-12v.html
2) Use of 2x [ 4 Cylindrcal Magnets & 1 plexiglass base
to mount the rotating magnets (available) ]
3) Simple, cheap & reliable RPM control with a N-Mosfet and a Potentiometer
https://www.youtube.com/watch?v=ipiDHD4YEMM
Test tube linear shaker:
Requirements:- 50 mm stroke
- 150 strokes/min shaking frequency
- 1 kg maximum shaking table load
For the definition of the crank, rod lengths (l1, l2 respectively):
- l1=25 mm, i.e. half the stroke length
- For the length of the rod, various values where tested, subjected to the kinematics restriction (l2>=l1). The values of the velocity in each case are shown below. For (l1=l2) the velocity [m/s] - crank angle [rad] function plot (figure 3.1) shows that the test tube table moves with high velocities only between {-pi/2+2·k·pi, pi/2+2·k·pi}, where k=0,1,2,... . That creates high inertial acceleration resulting in impulsive loads exerted at the joints and the motor.
Figure 2.1: velocity=function(cam_angle) for l1=l2.
With a little research [3.1] in combustion engine crank-rod mechanisms, we deduce that some standard values of the rod length resulting in reduction of the above described phenomenon, are 3-5 times the length of the crank. Chosing a rod length three times the crank length (l2=75 mm), results in a much more gentle velocity curve (Figure 3.2) and finite values for the acceleration curve (Figure 3.3).
Figure 2.2: velocity=function(cam_angle) for l1=3·l2. The max velocity value was calculated as 0.207057 [m/s].
Figure 2.3: acceleration=function(cam_angle) for l1=3·l2. The max acceleration value was calculated as 1.11604 [m/s^2].
From the maximum value of the accelerarion, we can calculate the maximum inertial force exerted at the table of total weight of 1 kg, being Fmax=1.11604 [N]. Multiplying the maximum force with the maximum velocity,results in the ideal maximum power consumption being Pi,max=0.2311 [W]. To be able to sustain a reciprocating frequency of 150 strokes/min in this mechanism, a steady angular crank velocity of n=75 [rpm] is needed. We chose a fitting dc motor for this purpose {n_max=72 rpm, P_max=7 W, V=12V}. After some early mechanical strength calculations the following crank-rod mechanism was designed:
Figure 2.4: Crank-rod linear shaking mechanism.
List of Components:
DC Motor 75 rpm
Crank-Rod
2 bent corners
2 flanged ball bearings MF85ZZ
2 axel
2 linear ball bearings
2 custom made threaded studs
1 adjustable linear bearing nest
Air Heating System
Most of the commercially available operational shaker-incubators do not clarify how the temperature control system works. However, there are basically two options:
1) Use of a PID (or similar) controller. One could also try to model the thermal system (resistor, capacities, sources, etc) but it is doubtful that this estimation would be anywhere near reality due to the complexity of such systems.
2) Activation of the heating system (in our case, the fan and the resistor) when the temperature falls below minimum (in our case 32 or 37 Celcius). The system operates until temperature reaches maximum (defined by the user) and then it is deactivated. When the temperature reaches its minimum allowed value again (because of thermal losses), the system is activated again, and so on.
The main difference between the two methods is, as implied above, that the first is constantly adjusting the temperature so that is does not diverge significantly from the set value. However, it is more complex than the second and it requires the definition of gains. In the second case we accept that there is going to be an important fluctuation of the temperature's value, but we end up with a simple operating system in return. We decided to use the second method of control, regarding the reasons mentioned above. There will be one thermocouples,
The selection of both the fan and the resistor is of great importance in this application. The reason for this is that the air flow that the fan produces needs to absorb the heating power of the resistor, otherwise the resistor will overheat and eventually melt. This, of course, will lead the whole system to failure.
The resistor that will be used in this application can produce 1500 W and it is made out of nichrome whose melting temperature is 1400 oC. The fan that was chosen for this project can provide 145 m3/h of air flow. The following calculations will show that system above can be balanced so that the resistor's temperature does not exceed nichrome's melting point.
In a first attempt to get a good estimation of how the heat transfer phenomena will affect the temperature of the resistance, the following assumptions were made. The first assumption was that the main phenomenon of heat transfer, in this application, is convection. And the second one was that the shape of the resistance can be modeled by a cylinder of the same diameter and length as the actual resistance. The second assumption, although not true, was made in order to approximate the heat convection factor. The heat convection factor is affected by many variables such as the relative speed between the liquid that moves around the resistance (in this case air) and the resistance itself, the kinematic viscocity of the liquid, the diameter of the resistance and the type of the flow (direction in which the flow approaches the resistance. So, it is easy to understand that the heat convection factor is very hard to calculate. However, the heat transfer for a flow around a cylinder is a well studied phenomenon and thus the equations that describe it were available to us.
Having all the above in mind , the following equations can be written:
P_resistor=Q ̇_convection⟹
1500W=m ̇_air∙C_p∙ΔT⟹
So, it is possible to calculate the air flow needed to achieve the desired air temperature int the incubator which is approximately 40oC:
m ̇_air=1500W/(1005 J/(kg∙°C)∙(40-15)℃)=0.059 kg/s⟹
Therefore, the velocity of the incoming air can be found from the following equation:
v_air=m ̇_air/(ρ∙Α_fan )=(0.059 kg/s)/(1.2 kg/m^3 ∙π∙((0.115 m)/2)^2 )=14.87 m/s
For this velocity, it is possible to estimate the convective heat transfer factor:
h=100 W/(m^2∙K)
Since all the above is found, the temperature of the resistor T_s is calculated:
Q ̇_convection=A_resistor∙h∙(T_s-15°C)⟹
T_s=1210℃
So, it is shown that the resistor will operate at a temperature which is well underneath its melting point and thus the selection of the fan is suitable. However, the model of the cylinder that we used in this approach cannot describe the reality in a very satisfactory way, since the temperature found was well above the expected values.
For this reason, a second approach was attempted. In this approach, the program Ansys Transient Thermal was used to get a more accurate value of the temperature of the resistance as well as the time in which this temperature is achieved. In order to get a better estimation, a better model of the resistance was needed. In that spirit, the shape of a cone with a hole around its axis was adopted. The assumption that convection is the major phenomenon taking place, though, was not altered. After the program was used the following results were introduced:
So, in these diagrams we can see that the temperature of the resistances is stabilized at 469.85 oC after 50 seconds. These results are much more appropriate to describe the result we expect and they show that the resistance is not in danger of overheating. However, it is important to notice that the inputs of the program were the power of the resistance and the heat convection factor. For the second input we used the value of the convection factor that we calculated in the previous approach. That means that this is only a good approximation and the necessity of an experiment that verifies our results cannot be overwritten.
Because the fan was larger than the resistor (120 mm diameter vs 40 mm), and because the computer fan used is not designed to provide axial flow, it is necessary to use a "tube" in order to focus the air flow on the resistor. Running a simulation on ANSYS CFX we found that:
Pressure right after the fan measures about to 650 Pascal which is equal to 66.281 mm H2O. However, our fan, according to the pressure-air flow curve, can provide up to 8 mm of pressure.
We could therefore conclude that this fan is not appropriate for this kind of application. However an experiment was conducted. We constructed a frame from a carton box, like the one we simulated, only square, and we activated the fan. There was not as much air flow as without the frame, but when the resistor was fed with current, the temperature quickly rose to 50 degrees, more than enough. The only problem was that temperature decreased afterwards as quickly as it rose, requiring activation of the resistor through the relay every 5-10 seconds to keep it about 35 degrees. As a final conclusion, we should keep in mind that this system works well for small amounts of time, without any danger of inflammation, with the drawback that it has to be activated about every 20 seconds (if we consider that the actual construction will be better insulated than our carton frame).
Electric subsystem
The main challenge concerning the design of the electrical subsystem is the proper isolation between the AC and DC components, as well as their protection from excessive currents. Special care should be taken to protect the Arduino board, which is the most fragile and expensive component.
The motors used in the magnetic stirring are presented on the following link (https://www.servocity.com/html/303_rpm_gear_motor.html)
Considering our system should have a single plug to power the motors, the Arduino board, the resistor, the fan and every other electronic or electrical component, it is clear that the use of a transformer is imperative. Input current (230V/50Hz) supplies the resistor an its fan before transformation to +12VDC, which is used to supply the motion system's motors and the Arduino.
Other components include a temperature sensor (DS18B20, http://grobotronics.com/temperature-sensor-ds18b20-el.html ) requiring +5V supply and a relay (https://www.parallax.com/product/27115) to control the resistor. A relay could also be used to switch the fan on and off, but we chose to keep the fan always functioning to provide better air circulation, which is another system requirement.
Safety precautions are taken to protect the system's integrity. There is a fuse in the main power adaptor as well as a master switch which can be used to shut down the entire system in case of malfunction. Measures should also be taken to ensure that the resistor can not be supplied unless the fan is functioning, to avoid inflammation. In the schematics that follow, switch A ensures that this does not happen. Switch B is used to deactivate temperature control in case it is required to stir something in room temperature.
Last, concerning current draw, it is calculated to about 0.5-0.6 A for the DC subsystem and about 7 A for the AC subsystem, plus the transformer AC current draw. The fuses should be chosen accordingly, also taking into account higher currents that occur when motors start rotating.
The circuit is pictured below:
Casing:
The main goals of the chamber's design are the following:
We came up with two main designs for the chamber:
1st design
In this solution, the only material and machining process necessary is laser cut of PMMA/Plexiglass material.
All the construction is assembled with screws, brackets and geometric special formations.
All the Motion System parts are placed on two slabs that are assembled together with four threaded rods.
After that, we can assemble the rest of the chamber with simple M5 screws and nuts.
In the back side, all the Thermal System Components are placed. Then, the cover is assembled using two simple hinges.
The last piece of the chamber is placed using the geometry formation showed in the next picture and can be screwed with 2 threaded rods as seen below:
This construction has the benefit that it is extremely easy to assemble, using just a screwdriver and a wrench but has the downside that it's heavy, around 10kilos.
2nd design
In the second solution, we replaced the most PMMA parts with two steel sheets. Each sheet's thickness is 0,5mm. This reduced the total weight of the chamber by 40% and made the assembly procedure even easier and quicker.
Download 3D-CAD Assembly here
The flat and bent sheets are the following:
The Motion and Thermal system controls are placed on the front face of the chamber.
The Thermal System components are placed inside the back of the chamber.
Under the chamber's bottom there are 4 rubber pads that can absorb the vibrations caused by the Motion and Thermal System.
The construction includes a microswitch connected with our arduino and is placed on the left side of the chamber.
The switch's purpose is to stop the stirring function when the cover is open.
List of Components:
2 Steel Sheets
4 Two-plane Brackets
2 Three-plane Brackets
4 Plexiglass Slabs
1 Handle
2 Hinges
4 Pads
4 Threaded Rods
8 M10 nuts
22 M5 screws
22 M5 nuts
24 Sheet Metal screws
Chamber's Total Cost:
The total cost of the brackets/pads/screws/nuts/hinges and handle is ~15euros
The plexiglass slabs are already in our disposal.
The price of the bent steel sheet is still unknown. It depends on the machine shop's offer.
Milestone 3: Prototype Fabrication
Shaking Mechanism Subsystem:
gif 3.1: Isometric view of mechanism function
Experiment of Magnetic Stirring:
Experimental setup:
1 acrylic base
4 cylindrical magnets
1 screw
Rotational motion of a drill
1 Erlenmeyer flask with 100 ml water
1 magnetic capsule
Experimental conclusions:
A magnetic stirrer with the existing magnetic setup will function well in 300 rpm, given that the rotational speed will be increasing gradually. The magnetic capsule rotates due to the magnetic field created by the four magnets, creating the needed vortex for the air to mingle with the water.
Final Assembly:
Figure 3.2: Complete Assembly in isometric view.
ReferencesResearch Papers1.2.
Web sites1. https://www.youtube.com/watch?v=x1pPdB7RlHQ2. http://web.mit.edu/2.75/fundamentals/FUNdaMENTALS.html (focus on chapters 3-7)
Books:[3.1] "Μηχανές εσωτερικής καύσης Ι" του Καθηγητή Κ. Δ. Ρακόπουλου σελ. 224[3.2] "Αρχές Μεταφοράς Θερμότητας και Μάζης", Ξενοφών Κακάτσιος, 2006, Εκδόσεις Συμεών
(c) 2015 Department of Mechanical Engineering, National Technical University of Athens.